HP Proliant ML350 G6 server Auto OFF problem

[Gowtham-mano]

I have two HP Proliant ML350 G6 servers in our plant. We are facing problem of  Auto OFF. This occured more than 6 times wtithin a span of 4 months. But when i check the UPS there is no trip or Power supply failure from its side and also i checked the voltage level at the server side its found ok. So please suggest any option to sort out this problem or  whether the problem may be in the SMPS of the server.

[waaronb]

Did the management log show any errors or entries when the server shut down?

I think the default for Proliant servers is to restore the previous "ON" state when it loses power... so if it was running when power was lost, it would turn back on again when power was restored. Unless you changed that, if the power failed the machine should have started back up again when the power came back.

Overheating would cause the thermal shutdown to kick in. When that happens, it shuts itself down but I think the machine will start itself back up after a while and see if it's cool enough. I haven't seen that happen to my servers in a while so I forget exactly how it works.

ASR (auto server restart) will activate if you have the HP agents installed. It checks for a heartbeat from the system and will restart if it doesn't see the beat in 10 minutes or whatever by default. But that's a restart, not a shutdown.

Other reasons for the system shutting down could be some hardware problem... memory, CPU, power supply, etc. The maintenance log usually shows what happened but if it was a sudden failure like a power supply failing, it wouldn't log that, but it might show an error when it does power back on... something like "power lost/power restored".

[Gowtham-mano]

The message wat I get by running Dcdiag after auto off is "improper shut down may be due to hardware crash or power failure " but I have hp z210 4pcs on the same UPS supply but it didn't get off.

[Jimmy Vance]

That is a message from the OS, what does the server IML, or iLO logs show?